nMASE: A Search Engine for Network Trace

A Search Engine which records and analyzes network activity and provides the user with ranked results through a web interface.

Computer networks have been largely adopted in small and big settings in corporate and residential environments. With explosion of mobile traffic and ever increasing use of software defined networking, network traffic monitoring holds a key position in detecting network errors and security failures. Monitoring big networks has become a herculean task with having to deal with huge magnitude of data which is prohibitive for proper leverage considering the time and space required to exploit it. Much of the current work in this domain is focused on techniques to efficiently analyze and classify network traffic. Rule based classification or filtering of traffic is the first major step in any analysis procedure so that analysis is performed on focused set of packets. However defining, constructing and application of such rules involve considerable level of difficulty. In this project, we address this issue.

We believe that searching for relevant content in network trace should be as easy as searching for relevant web pages through search engine. In this paper we touch upon this much neglected area by introducing concept of network search engine and ranking of network traffic. “nMASE- A Search Engine for Network Trace” is built upon Boolean model of information retrieval and is capable of finding relevant content from a huge network trace based on simple natural language queries. This leads to significant reduction in the time that network administrators spend in scanning through colossal network activity.

• Date: Sept 2014 - Feb 2015
• Languages: C, Python, HTML/JS
• Github link: https://github.com/vipulchaskar/nMASE
• Paper link: https://www.ijsr.net/archive/v5i8/ART2016907.pdf
• Domain: Networks

Close

Processor Pipeline Simulator

Developed as part of CS520

Microprocessor instruction processing pipeline simulator (Developed as part of CS520) This program simulates execution of an out-of-order, multi-FU processor pipeline for the given stream of (a subset of RISC-like) assembly instructions. The simulator has following features:

• Load-Store instructions to interact with memory
• Load-Store instruction queue to serialize access to memory with support for LOAD-Forwarding and LOAD-Bypassing
• Reorder buffer (ROB) for inorder writes to registers
• Physical registers and register renaming
• Issue Queue for waiting instructions
• Data forwarding from FUs and memory unit to Issue Queue and LSQ
• Support for conditional and unconditional branch instructions
• Support for flag-dependent instructions

• Date: Fall 2017
• Languages: Java
• Github link: https://github.com/vipulchaskar/Pipeline-Simulator
• Domain: Microprocessor

Close

Distributed, Fault-tolerant, HA Key-Value Store

Developed as part of CS557

An eventually-consistent distributed key-value database like Cassandra. Eventual consistency is implemented by read repair and hinted handoff. Uses google protobuf to define messages to be sent. Supports 4 replicas at a time but can be modified.

• Date: Fall 2017
• Languages: Python
• Github link: https://github.com/deshpandekartik/Cassendra-type-Distributed-NoSQL-database
• Domain: NoSQL

Close

Equation Solver as a Service

Developed as part of CS571

A REST API which accepts a picture of a linear equation(s) or mathematical expressions, extracts the equation from picture, solves it, and returns the result. The constraint given was that we had to use 3 different programming paradigms.

We used Java for front end since Java is widely used in web back-end technologies and there are plenty of frameworks available like spring. We used python to extract equation from the provided image (OCR). This is because Python has plenty of libraries available for such tasks. We used Haskell to solve this extracted equation because functional languages like Haskell are primarily used for mathematical computations.

• Date: Fall 2017
• Languages: Java, Python, Haskell
• Category: Mathematics

Close

Distributed banking application - Global snapshot

Using Chandy Lamport Global Snapshot Algorithm

Developed a set of branches of a bank running on different machines. All are provided a fixed initial balance by another program called Controller. These branches then keep sending a random amount of money to a randomly selected branch after random intervals. The controller periodically takes snapshot of all the branches. This snapshot includes balance of all the branches and the money currently in transit. The sum of all the branch balances and money in transit must match the initial balance that was provided by Controller.

• Date: Fall 2017
• Languages: Python
• Github link: https://github.com/vipulchaskar/Chandy-Lamport-Snapshot-Algorithm
• Category: Distributed Systems

Close

Company Finance Management System

Undergrad academic mini-project

A CRUD application for handling and tracking financial transactions of an organisation. Can keep track of daily expenses, inventory purchases, sources of income and can generate corresponding logs and show relevant graphs.

• Date: Fall 2014
• Languages: VB.NET, Oracle database
• Category: CRUD

Close